From measurement to mitigation: a historical review of livestock emission research in the U.S. Southern and Central Great Plains

1 Introduction

The livestock industry represents a cornerstone of U.S. agriculture, contributing $268 billion to the economy in 2024 (USDA, 2026a) and producing over 100 billion pounds (45 million metric tons) of red meat and poultry combined (USDA, 2026b). Over the past decades, industry has experienced significant growth, particularly in concentrated animal feeding operations (CAFOs; Lee et al., 2025). There is an extensive and significant geographic footprint of varying CAFO densities across numerous rural counties throughout the country, including a prominent presence visible across the Great Plains region (Figure 1). The production of red meat and broilers increased by 50 and 180%, respectively, from 1990 to 2025 (Figure 2). More production reflects more animals raised in CAFO, and thus a greater emission burden. However, that expansion has brought considerable environmental challenges, as CAFOs can affect air quality through emissions of ammonia (NH3), hydrogen sulfide (H₂S), odor, particulate matter (PM), volatile organic compounds (VOCs), and greenhouse gases (GHG; Pohl and Lee, 2024; Saifuddin et al., 2025; Zhu et al., 2015). Those concerns have prompted growing public and regulatory scrutiny of emissions from livestock operations (Ashwood, 2021; Cole et al., 2003; Vogeler, 2021), with pressure intensifying as rapid urbanization in rural areas has brought residential communities closer to livestock facilities, resulting in increased complaints about nuisance from CAFOs (Bokowa et al., 2021; Jerez et al., 2013; Mckittrick et al., 2025). Recent nationwide spatial assessments have amplified those concerns, demonstrating elevated PM_2.5 (PM with diameter < 2.5 μm) concentrations in vulnerable communities adjacent to CAFOs (Chamanara et al., 2025).

Contrary to the perception of industry inaction, the agricultural sector has been engaged in a proactive, federally backed effort to address these exact emissions for decades. The regulatory and research response to these challenges can be traced to 1996, when the U.S. Congress recognized that published data on agricultural operations were not sufficiently reliable and established the Agricultural Air Quality Task Force (AAQTF), which subsequently recommended up to $65 million in research funding per five-year period in the early 2000s (Parnell and Parnell-Molloy, 2002). Given that substantial investment in livestock emission research, it is critical to understand how effectively these funds have been used to drive innovation and develop solutions to address these pressing environmental concerns, particularly in the U.S. Great Plains region where livestock production accounts for more than half of agricultural industry receipts (Klemm and Briske, 2021).

Emissions quantification serves as the main pillar of livestock emission studies, establishing the scientific basis for why research must be conducted and providing the data necessary to assess the magnitude and severity of impacts. Emission quantification studies are essential for establishing emission factors (EFs), which have multiple uses such as multiscale modeling, regulations development, and permitting (Habib et al., 2024; Liu et al., 2017a). A significant challenge in developing EFs is that values vary considerably across geographical regions, climates, and measurement techniques. Furthermore, emission quantification is complicated by measurement protocols borrowed from other industrial contexts and not designed for the open, weather-driven conditions of agricultural operations, where wind, temperature, and air movement strongly affect measured emissions (Jerez et al., 2006; Ma et al., 2024; Moshayedi et al., 2023).

Beyond providing EF for regulatory purposes, emission quantification studies serve to identify specific conditions under which emissions are greatest, such as particular locations or operations within livestock facilities (Hafner et al., 2026; Mukhtar et al., 2009) or specific seasons or times of day (Gómez-Garrido et al., 2026; Rodrigues et al., 2024; Todd et al., 2011b). Additionally, these studies can identify both natural phenomena and management interventions that reduce emissions, such as rainfall or water application for PM (Bonifacio et al., 2011; Guo et al., 2011b). These insights enable researchers to focus future studies strategically and target the development of abatement solutions where they are most needed, underscoring the growing need for precision approaches in which site-specific emission data can inform targeted, facility-level air quality management strategies.

Effective emissions regulation depends critically on the scientific validity and reliability of measurement methods, yet this has been a persistent source of controversy in CAFO emission research. One of the most common concerns was that the regulatory framework was not in place, and/or sufficiently scientific and lacked reliability. Specific sections of the U.S. Environmental Protection Agency (EPA) regulations, such as AP-42, the Compilation of Air Pollutant Emission Factors, have been frequently debated, updated, and subjected to controversy since the early 2000s (Goodrich et al., 2003). The EFs within AP-42, which serve as the basis for estimating emissions from CAFOs, have been particularly scrutinized. Early work showed that factors developed in industrial settings were often inapplicable to agricultural operations (Shaw et al., 2004; Shaw et al., 1998). More recent work highlights a further limitation: fixed EFs do not transfer well across locations, because differing regional conditions and management practices drive emissions in ways a single factor cannot represent (He et al., 2026; Sommer et al., 2022). When those scientific inadequacies are built into regulations, the effectiveness and even-handedness of the resulting regulatory apparatus are open to question.

Alongside these challenges emerged opportunities, as measurement technology for agricultural operations developed rapidly, offering new avenues to quantify emissions with higher accuracy and resolution. This was particularly important for odor emissions, as odor measurement presents unique difficulties given that odorous gases typically occur at very low concentrations and are often accompanied by hundreds of ancillary, non-odorous compounds (Bulliner et al., 2006). Similarly, major improvements were made for measurements of NH₃ and GHGs with, for example, automated flux chambers for area sources (Parker et al., 2021), open-path lasers (OPL) (Todd et al., 2015) tuned for target gases, cavity ring-down spectroscopy (CRDS)-, Fourier-transformed infrared (FTIR)- and photoacoustic-based analyzers (Leytem et al., 2013) for continuous measurements in rural environments. These needs and opportunities constitute the most challenging pillar of livestock emissions research: measurement technology, which seeks to develop reliable methods to quantify the substances that contribute to pollution emissions.

The ultimate goal of livestock emission research is to develop practical solutions that enable producers to mitigate emissions. These solutions span a diverse range of interventions, including altering animal diets to reduce emission precursors (Ahmed et al., 2024; Malyugina et al., 2025; Spiehs et al., 2012b), using biochar to filter air within animal facilities facility (Chen et al., 2009; Choi et al., 2025; Park et al., 2025), and applying additives to manure storage to suppress emissions (Cai et al., 2007; El Bied et al., 2023; Kavanagh et al., 2021). The effectiveness of abatement technologies is typically evaluated based on two critical criteria: the potential reduction in emissions they can achieve (Maurer et al., 2016) and their economic viability for producers (Cantillon et al., 2024; Luke and Tonsor, 2024; Parker et al., 2004). This research area represents the translation of scientific knowledge into actionable solutions in which cost–benefit relationships can be quantified.

This historical review examines livestock emission research conducted in the U.S. southern and central Great Plains region, particularly as it ties to the context of the AAQTF charter. The primary objective is to report the progress of livestock emission research and explore how different research domains interact and inform one another. The review is structured around three main research pillars: (1) Measurement Technology, (2) Emission Quantification, and (3) Abatement Solutions. By analyzing these three pillars in conjunction, this paper aims to (I) explore how the interdependence among these research domains has shaped the regional understanding of emissions from livestock agriculture and (II) identify knowledge and technology gaps that limit real world impact.

2 Literature search and screening methodology

This study is a structured historical review of livestock-emission research in the U.S. southern and central Great Plains, focused primarily on the federal research era of 1998–2017 and supplemented with selected recent literature for context. Literature was identified via a comprehensive search of the Web of Science Core Collection database (Clarivate, 2025), conducted on 30th September 2025, for the period between 1998 and 2017. Consistent with the historical objective of characterizing the research legacy of the AAQTF, the search strategy was intentionally centered on the lead investigators and institutions associated with that regional research effort. The search strategy employed author-based and keyword-based filters to identify relevant publications. This list comprised the lead investigators named on, or directly collaborating within, the AAQTF-linked federal appropriations to the regional land-grant and USDA-ARS programs, spanning Texas A&M AgriLife, Kansas State University, USDA-ARS (Bushland and Clay Center), and Iowa State University (see Supplementary material S1 for details). This targeted approach was cross-referenced with broad keyword filters to maintain a comprehensive scope while ensuring the capture of foundational technical reports and peer-reviewed outputs. Keyword searches combined Boolean terms related to livestock production systems (‘livestock’, ‘animal’, ‘swine’, ‘poultry’, ‘cattle’, ‘feedlot’) with terms (“AND”) related to air emissions and pollutants (‘air quality’, ‘emissions’, ‘odor’, ‘ammonia’, ‘NH3’, ‘manure’, ‘waste’, ‘particulate matter’, ‘VOC’, ‘volatile’). This identification phase yielded 358 initial records, which were subsequently screened based on title and abstract relevance to CAFOs. Duplicated entries were removed. Exclusion criteria were strictly applied to filter out literature not directly related to agricultural systems, such as studies on game animals, zoo species, or general urban pollution. White papers and older conference proceedings where full text could not be retrieved were also removed. Following the removal of ineligible records, 190 studies were retained for full retrieval and intensive analysis. The final research database was categorized into a thematic matrix representing the foundational pillars of the review: 46 studies focused on measurement technology, 77 on emission quantification, 52 on abatement solutions, and 15 reviews. Each study was further analyzed based on specific criteria, including livestock type and emission categories, and sub-categorized by abatement approaches. In addition to the 1998 to 2017 literature, a limited number of more recent publications were added to provide updated context on the current state of technology.

3 Measurement technology

Measurement technologies for livestock facilities can be broadly grouped based on their targeted pollutants: VOCs as the primary contributors to odor nuisance, PM, and gases concerning GHG emissions, hydrogen sulfides (H₂S, an odorous gas), and NH₃. Among these categories, the majority of measurement technology studies focused on odor and VOC characterization due to multiple persistent challenges. First, hundreds of distinct chemical compounds can be associated with livestock operation odors, making comprehensive identification and quantification particularly complex. Second, methods to quantify odor and link specific chemical compounds to human olfactory perception remained underdeveloped throughout the early 2000s, creating fundamental gaps in understanding what constitutes odor nuisance. Third, odor-causing compounds are often present at extremely low concentrations, frequently in the parts per billion range, making sample collection, storage, and analysis technically demanding without degradation or loss of target analytes.

PM measurement faced a distinctly different set of challenges rooted in methodological incompatibility. Standard PM monitoring protocols were originally designed for urban and industrial environments where particle size distributions, and emission mechanisms differ from agricultural operations. When these established regulatory methods were applied to agricultural settings, they exhibited considerable measurement errors and often required modification for agricultural applications.

3.1 Methodological challenges in agricultural air quality monitoring

3.1.1 Capturing the invisible: overcoming odor and VOC measurement barriers

3.1.1.1 Odorous air sampling collection, storage and recovery biases

Early odorous air sampling technologies in the 2000s faced significant challenges with sample degradation and storage (Parker et al., 2010b). Solid sorption tubes, which used thermal desorption to release captured compounds for analysis, often degraded or contaminated VOC-targeting samples during analysis process (Koziel et al., 2005). This limitation led researchers to adapt solid-phase microextraction (SPME) from the food and flavor industries (Koziel et al., 2001; Razote et al., 2002). SPME uses fiber coated with specialized sorbent materials placed in the air or headspace above a sample to extract VOCs. Razote et al. (2002) tested two SPME fiber coating types in swine houses and found that Carboxen/polydimethylsiloxane (CAR/PDMS) fibers achieved reproducible measurements of odorous VOCs with relative standard deviations ranging from 4 to 17%.

Beyond sampling and sample preparation methods, sample integrity in storage presented equally serious problems. The widely accepted practice of collecting odor samples in commercial Tedlar bags significantly biased results through background chemical interference and rapid volatile loss. Sample recovery rates can be reduced to as low as 0 to 14% after 7 days’ residence in the bags (Koziel et al., 2005; Parker et al., 2010b). Koziel et al. (2005) comprehensively tested eight different sampling media, including various commercial Tedlar bags from different manufacturers, PET bags, Tenax TA sorbent tubes, and SPME fibers. The study confirmed SPME as the most reliable method with a mean recovery rate of 98.3% after 24 h, followed by Tenax TA sorbent tubes at 94.8%. Importantly, the research identified specific residual compounds leaching from different bag materials, such as phenol and N, N-dimethyl acetamide, that could interfere with olfactometry measurements.

To address these containment issues, researchers developed novel metallized fluorinated ethylene propylene (FEP) bags in the mid to late 2010s. Zhu et al. (2015) compared these new FEP bags against conventional Tedlar bags and found that FEP bags not only achieved higher VOC recovery rates but also exhibited significantly lower background interference from bag materials. Meanwhile, researchers optimized passive and time-weighted average (TWA) SPME configurations (Haddadi et al., 2025) to detect parts per billion concentrations in harsh, biosecure agricultural environments without damaging the delicate fibers (Koziel et al., 2017; Tursumbayeva et al., 2019; Zhu et al., 2015).

3.1.1.2 Sensor inadequacies and calibration hurdles

VOCs are among the primary compounds responsible for odor generation (Razote et al., 2002). Before VOC measurements in livestock production settings could be reliable, researchers first needed to create accurate calibration standards, which proved surprisingly difficult. Establishing standard gas mixtures for volatile fatty acids (VFAs, playing important role as odor contributors) was highly challenging because these compounds are extremely reactive and polar, making them unstable in conventional calibration gas cylinders. Spinhirne and Koziel (2003) developed methods using permeation tubes to generate stable VFA calibration gases in moving air, providing a critical foundation for subsequent odor detection experiments in agricultural settings.

Even with proper calibration standards, commercial sensors for odor-relevant compounds presented significant challenges. In the early 2010s, attempts to automate VOC monitoring using commercial electronic nonmethane hydrocarbon (NMHC) analyzers largely failed in agricultural applications (Trabue et al., 2013). These analyzers were based on U.S. EPA Method 25A (USEPA, 2017), which was designed for industrial and urban environments where VOCs are predominantly nonpolar hydrocarbons. In contrast, VOCs from livestock facilities contain high proportions of polar, oxygenated compounds. Trabue et al. (2013) evaluated three commercially available NMHC analyzers and found they drastically underreported true VOC concentrations, missing alcohols by over 70%. The study concluded that if the VOC profile contains significantly high levels of polar oxygenated compounds, these commercially-available NMHC analyzers cannot be reliably used for monitoring of agricultural emissions without substantial modifications or corrections.

3.1.1.3 Correlating odor chemistry with human perception and air dispersion

Purely chemical mass & concentration measurements do not capture human perceived smell, creating a challenge in agricultural odor research. To address this gap, researchers developed multidimensional gas chromatography–mass spectrometry coupled with olfactometry (MDGC-MS-O), which allows trained human panels to sniff effluent (separated compounds) from the chromatographic column and simultaneously correlate chemicals’ identity and abundance with perceived characteristic and offensive odor (Bulliner et al., 2006; Laor et al., 2008; Wright et al., 2005; Zhang et al., 2010). In the early 2000s, there was a critical lack of comprehensive databases associating the characteristic and specific odors from CAFOs with their chemical compounds (Laor et al., 2008). Subsequent studies systematically identified compound profiles associated with different livestock operations, including swine (Akdeniz et al., 2011; Cai et al., 2009; Koziel et al., 2010; Lo et al., 2008), beef (Cai et al., 2006), and poultry facilities (Koziel et al., 2017; Trabue et al., 2013).

However, a fundamental challenge emerged regarding best practices for measuring agricultural odors, leading to two distinct methodological philosophies (Trabue et al., 2011). The first approach argues that odor is a complex sensory experience sometimes greater than the sum of its individual chemical parts due to synergistic, antagonistic, and potentiating interactions between and among compounds. This method relies on whole air sampling and sensory evaluation by trained human panelists to quantify odor intensity and character, frequently employing dynamic dilution forced-choice olfactometry (DDO) to evaluate field air samples and determine dilution-to-threshold (DT) concentrations (Jacobson et al., 2008). While this captures the true human experience of odor detection (yes/no), the sensory approach faces severe logistical limitations, particularly the rapid sample degradation and contamination and loss of agricultural odorants when stored in standard collection containers. In addition, this approach says nothing about the offensive character of odors evaluated, i.e., resulting odor DT ‘concentration’ could be the same for air samples being very offensive and pleasant in odor character. The latter challenge can be misleading and severe when assessing downwind impact in complex airsheds with multiple possible contributing sources (Wright et al., 2024; Wright et al., 2021).

Conversely, the second methodological approach posits that complex odors can be effectively characterized by isolating and quantifying only the key handful of chemical constituents that impart the characteristic offensive smell of odor. Rather than relying on DT and subjective human olfaction for measuring odor concentration only, this approach uses a combined chemical and sensory approach via advanced analytical instruments to identify specific VOCs responsible for the overall smell in inherently complex airshed of multiple sources characterized by highly variable environment (both spatially and temporarily). This approach was recently embodied in the ASTM D8623-25 standard practice (ASTM, 2025). By applying advanced SPME and MDGC-MS-O methods in the mid to late 2000s, researchers successfully tracked specific VOCs downwind and identified p-cresol as the dominant compound carrying the characteristic cattle and swine odor at distances up to 16 km from the source (Koziel et al., 2009; Koziel et al., 2006; Wright et al., 2005). Researchers using this chemical approach have identified up to 295 distinct compounds emitted from swine manure, attempting to isolate a small percentage of character-defining odorants that dominate the overall odor profile (Lo et al., 2008). Moreover, simultaneous chemical and olfactometry analyses of dust filters revealed that PM acts as a major carrier for odor, with characteristic swine odorants preferentially binding to smaller-sized PM fractions (Cai et al., 2009), signifying greater airborne odor-carrying potential for fine PM. Taken together, this insight offered an important implication for odor investigations and mitigation, i.e., the need to focus on identifying the key compounds imparting the characteristic offensive odor perceived downwind from the source and prioritizing removal.

3.1.2 Measuring agricultural PM: when standard methods fail

3.1.2.1 Incompatibility of dust regulatory equipment with agricultural environments

Early PM monitoring efforts in agricultural settings revealed incompatibilities with established regulatory methods developed for urban and industrial environments (Buser et al., 2001). Standard Federal Reference Method (FRM) samplers, designed to measure ambient urban PM, produced massive concentration measurement errors of up to 343% when applied to agricultural operations (Wanjura et al., 2005). This substantial error occurred because PM emitted by agricultural operations, such as cotton gins and CAFOs, has a significantly higher mass median diameter (MMD) compared to urban and industrial PM. The inlet designs and size-selective characteristics of FRM samplers were optimized for the smaller PM size distributions typical of combustion sources and urban environments, rendering them unsuitable for capturing the coarser agricultural dust without major modifications.

3.1.2.2 Limitations and disagreement of continuous dust monitoring technology

Even when researchers attempted to use continuous monitoring technology, significant challenges emerged regarding reliability and consistency. Multiple reference methods existed to measure PM concentrations, yet their readings were often inconsistent with one another. Continuous PM monitors such as the tapered element oscillating microbalance (TEOM) suffered in the humid, extreme barn environments characteristic of CAFOs, requiring constant maintenance to prevent rapid filter plugging (Guo et al., 2009; Heber et al., 2006; Wanjura et al., 2008; Jerez et al., 2006). Most studies reported that TEOM monitors consistently read lower PM concentrations compared to the ‘gold standard’ gravimetric method (Wanjura et al., 2008; Jerez et al., 2006), with moisture appearing as the main culprit, as measurement variability increased substantially with relative humidity that is susceptible to continuous change (Guo et al., 2009; Heber et al., 2006; CLi et al., 2012).

Gravimetric sampling itself faced limitations despite being the reference standard. These samples required filters to be deployed for extended collection periods, typically 3 to 9 h, before retrieval and laboratory processing to calculate PM concentration (Wanjura et al., 2008). This made data collection extremely labor intensive and provided poor temporal resolution, which proved unsuitable for capturing important diurnal emission patterns such as evening dust peaks commonly observed at livestock facilities (Auvermann et al., 2010; Marcillo and Auvermann, 2025; Peanusaha et al., 2026). The trade-off between accuracy and temporal resolution remained a persistent challenge throughout the 2000s.

3.1.2.3 Advanced tools for dust source determination

Traditional PM samplers could only measure dust mass in broad aerodynamic-sizing categories, providing little to no useful information about emission sources. Understanding where PM originated became critical for developing targeted mitigation strategies. Resolving the origin of agricultural dust emissions required advanced, non-destructive analytical tools such as Neutron Activation Analysis (NAA) and single particle Raman microscopy to physically trace whether particles came from animal manure, feed materials, or unpaved soil surfaces (Huang et al., 2013; Lange et al., 2009). These sophisticated techniques analyze the elemental composition or molecular structure of individual particles, creating unique chemical fingerprints for different source materials. For instance, Huang et al. (2013) used these methods to determine that most feedlot dust originates from dried manure and internal unpaved roads rather than feed materials, providing crucial insights for prioritizing dust control measures at CAFOs.

3.1.3 Quantifying gaseous emissions in harsh environments

3.1.3.1 High logistical costs, labor, and biosecurity constraints

Most advances in gas measurement technologies emerged from poultry industry research, driven by the necessity to measure emissions in harsh environments while managing constraints related to cost, labor, and biosecurity. The harsh air inside poultry operations, characterized by high NH₃ concentrations, elevated humidity, and suspended PM, can exceed the operational limits of standard analytical equipment. In the early to mid-2000s, researchers attempted to protect sensitive gas analyzers by housing them in specialized mobile laboratories, but this approach proved prohibitively expensive, costing $125,000 to $200,000 per unit, while transporting equipment between facilities created major biosecurity risks of disease transmission (Gates et al., 2005; Xin et al., 2003).

Labor intensity remained a persistent challenge throughout emissions-based research. Traditional non-flow-through, non-steady-state chambers used for measuring GHG fluxes in the late 2010s were extremely labor-intensive, requiring researchers to manually collect gas samples over 30-to-60-min periods before transporting them to laboratories for separate analysis (Parker et al., 2017a; Todd et al., 2011a). While some researchers investigated whether reducing sampling intervals could lower data collection costs, Liang et al. (2006) found that measuring less frequently than biweekly in poultry layer houses led to significant emission underestimation. This finding prompted others to focus on engineering solutions, such as standardized chamber designs (Parker et al., 2013a; Parker et al., 2013b; Parker et al., 2010a) and equipment modifications, to reduce measurement errors rather than simply reducing sampling frequency. Parker et al. (Parker et al., 2013a; Parker et al., 2013b) developed a standardization method for comparing readings from portable flux chambers using water evaporative flux ratio correction factors (EPRCF) to improve the accuracy of gas flux measurements.

Field measurements presented additional complications beyond controlled barn environments. Measuring gas emissions directly from outdoor waste lagoons proved difficult because surrounding trees and crops created highly variable flow fields, and lagoons were often too small to establish an equilibrated microclimate necessary for accurate flux measurements using micrometeorological techniques (Eve et al., 2014; Ro et al., 2013). Open-path tunable-diode lasers were proposed as a solution for open lagoons, such as the NH₃ and CH₄ work by Todd et al. (2011b) on dairy lagoons in eastern New Mexico.

3.1.3.2 Inaccuracies in emission rate calculations for animal facility

Accurately quantifying gas emissions required not only precise concentration measures but also reliable determination of building ventilation rates (Casey et al., 2007; Gates et al., 2004; Hoff et al., 2009). With respect to fully enclosed housing systems, measuring gas or PM concentration is only half the equation; the emission rate depends equally on knowing how much air carries those pollutants out of the facility at all times. Barn ventilation rates programmed for heat and moisture removal vary instantaneously but generally follow a characteristic daily and seasonal pattern. Those patterns affect the measurements requirements to capture that diurnal and seasonal variability, and therefore emissions estimates and their representativeness. In the early 2010s, researchers discovered that relying on published manufacturer fan performance curves (and therefore, ventilation rates) caused them to incorrectly overestimate actual NH₃ emissions by 13.6 to 26.8% (Casey et al., 2010). Manufacturer curves typically represent fan performance under ideal laboratory conditions with new equipment, clean air and optimal static pressure, which rarely match the dusty, high-humidity reality of livestock barns in which fans may interfere with each other, and their efficiency degrades over time. This finding revealed great variability across different measurement methods and demonstrated that emission quantification results could be significantly affected by the resources availability, but also by operator’s skill, experience, and choice of airflow determination technique. Furthermore, the estimation of barn ventilation rates is more complex when the barn is operated as a hybrid (part-time) mechanically ventilated or full-time naturally ventilated. The use of tracers, either non-naturally occurring (SF₆) or abundantly present (e.g., CO₂), has been proposed with some success, the latter requiring assumptions related to animal activity and emission status.

4 Emission quantification

Quantification studies in agricultural air quality research serve multiple critical purposes that have evolved to address both fundamental knowledge gaps and practical application needs. The most foundational purpose is to establish baseline EFs and pollutant concentrations for livestock operations. Such studies were conducted primarily due to the lack of comprehensive inventory data, with many cases where emissions of specific pollutants had never been quantified before (Li et al., 2011). Moreover, researchers quickly recognized that EFs could not be applied universally across different environmental conditions, animal housing systems, or management practices, requiring facility-specific or region-specific characterization (Bonifacio et al., 2012; Topper et al., 2008; Waldrip et al., 2014; Waldrip et al., 2013). The distribution of quantification studies across states (n = 69) is presented in Figure 3. Texas predominantly focused on cattle and dairy, while Nebraska, Kansas, and Iowa demonstrated greater diversity with studies spanning beef, dairy, swine, and poultry.

Beyond establishing baselines, quantification studies evaluate how environmental variables influence emissions and identify temporal trends in emission patterns. Understanding diurnal, seasonal, and condition-dependent variations provides a comprehensive picture of emission dynamics and helps identify peak emission periods that may require targeted management interventions. Furthermore, quantification studies demonstrate that temporal averaging, such as over 24 h periods, can obscure high emission events that disproportionately affect local air quality.

Quantification studies also serve to validate emission models, including atmospheric dispersion models and other model-based approaches for estimating EFs. This validation is particularly critical because EFs derived from different models can differ significantly, and methods suitable for enclosed facilities may not accurately represent emissions from open systems such as feedlots and open-lot dairies (Bonifacio et al., 2013a).

Some quantification studies focused on testing newly developed measurement technologies and generating datasets that expand available reference data. These efforts increased data availability for cross-study comparisons (Gonzales et al., 2011) and contributed to building chemical libraries for odor-causing VOCs and other target compounds (Bereznicki et al., 2012; Lo et al., 2008). Finally, applied quantification studies evaluate abatement solutions and facility design modifications, comparing the effectiveness of different mitigation strategies tested across various settings to identify high-potential interventions for reducing emissions from livestock operations.

4.1 Stages of emission quantification studies

4.1.1 Baseline development and condition-specific emission factors

Many early quantification studies were commissioned because agricultural sectors had almost no baseline emission data, leaving the industry vulnerable to regulatory criticism based on unverified estimates or questionable estimation procedures. For example, despite the large number of cattle raised in the southern and central Great Plains, there were virtually no data on the extent of NH₃ and H₂S emissions prior to these foundational studies (Baek et al., 2003). Furthermore, actual emission data for critical VOCs and GHGs from major sources like dairies remained exceptionally sparse (Sun et al., 2008).

Because concentration is simply estimated from a raw measurement of ambient air, researchers recognized the need to develop specific mathematical methodologies to convert these basic concentration readings into quantities scaled for facility throughput, EFs, required for legal permitting and national inventories (Lacey et al., 2003; Redwine et al., 2002). In enclosed, mechanically ventilated buildings, researchers established that this conversion is achieved by taking the measured gas or PM concentrations and multiplying them by the building’s measured exhaust air ventilation rate to calculate an overall mass emission rate (Hoff et al., 2006; Predicala and Maghirang, 2003; Redwine et al., 2002). Emission rates reported on dry standard conditions basis are crucial for cross-comparisons. Likewise, defining ‘standard conditions’ is also important as the EPA definition (USEPA, 1999) of standard temperature (25 °C) and pressure (101,325 Pa) can be interpreted differently, for example referring to the strict physics standard (0 °C and 100,000 Pa). Likewise reporting emissions based on at least several metrics such mass/time (M/t), M/t/head, M/t/live weight, M/t/animal unit (AU), M/t/emitting area, M/t/ barn or pen space per animal, etc., is also necessary for tabulating and comparing results from studies and developing EFs.

For open-lot facilities where air moves freely, researchers established that they must first calculate the emission flux (M/t/emitting area) using complex micrometeorological methods like reverse dispersion modeling or flux-gradient techniques, and then mathematically translate that flux into a per-head EF by multiplying it by the total pen area and dividing by the number of cattle (Bonifacio et al., 2012, 2015a; Todd et al., 2008). Ultimately, establishing these specific factors allowed researchers to report emissions inventory in terms of total mass per animal (or AU) over a specific and representative production cycle, such as defining an EF as 1.3 g of PM₁₀ (PM with diameter < 10 μm) per bird over a 7-week grow-out period, or calculating a cumulative emission of 141 g of NH₃ per turkey over a 108-d grow-out (Lacey et al., 2003; Li et al., 2011).

Though these foundational studies established critical baseline data, significant challenges remained in determining accurate and representative EFs. Emissions can vary drastically among geographical regions. Figure 4 illustrates the variability in NH₃ EFs derived from different livestock facilities across North America, with particular focus on the U.S. and Canadian Great Plains. This spatial variability underscores both the challenge of developing reliable, region-specific EFs and the motivation behind the large number of emission-quantification studies conducted during this period to address these knowledge gaps.

Determining accurate EFs proved crucial because these values serve as the basis for calculating several important regulatory metrics, most notably the potential to emit (PTE). The PTE represents the maximum capacity of a facility to emit a pollutant under its physical and operational design, and regulatory agencies use this calculated value to determine whether a CAFO should be classified as a major source polluter, triggering substantially more stringent permitting requirements, abatement technologies and threatening the imposition of emissions fees.

4.1.2 Temporal dynamics, environmental and meteorological drivers of emissions

4.1.2.1 Effects of temporal variability

Researchers recognized that emissions are highly dynamic rather than static, making the study of these trends critical because generalizing from single and random point measurements leads to inaccurate assessments of actual emission profiles. For example, odor emission rates were proven to vary significantly between seasons, across different days of the week, and even among different discrete locations on a single anaerobic manure pond, demonstrating that far more samples over longer duration are required to adequately characterize emissions than previously thought (Hudson et al., 2007b; Hudson et al., 2007a).

Daily diurnal patterns emerged as a major influence on gas, PM, NH₃ and odor concentrations. Specific compounds like VFAs frequently peaked at particular times of day, such as early morning and late evening (Trabue et al., 2011), proving that continuous, time-resolved monitoring is essential for capturing accurate emission profiles rather than relying on snapshot measurements that might miss peak emission periods (Auvermann et al., 2010; Peanusaha et al., 2026; Rhoades et al., 2010).

4.1.2.2 Characterizing environmental and seasonal emission drivers

Understanding the environmental drivers of emissions represented an important component of quantification research because weather events and meteorological conditions could substantially influence both the magnitude and chemical composition of livestock facility emissions. One significant motivation was the need to understand how rainfall triggers transient spikes in GHG production. For instance, researchers demonstrated that a single simulated rainfall event could trigger two distinct episodes of nitrous oxide (N₂O) production from manure-affected surfaces, with the delayed second episode accounting for up to 91% of the cumulative N₂O emissions over a 45 d period (Aguilar et al., 2014a; Parker et al., 2017b).

Pen surface moisture conditions also altered the chemical profile of odor emissions. Researchers tracked surface moisture because dry surface conditions produced the largest VFA and aromatic compound emissions, while wet or saturated conditions shifted the chemical profile to produce spikes in sulfide emissions (Woodbury et al., 2015). Understanding moisture dynamics also proved relevant for PM management, as studies demonstrated that maintaining pen surface water content of at least 20 to 30% effectively reduced PM₁₀ and total suspended particulate (TSP) emissions by as much as 60% (Bonifacio et al., 2013b, 2015a; Guo et al., 2011b).

Quantifying seasonal emission dynamics became necessary because temperature directly influences biological decomposition rates and the physicochemical processes governing pollutant volatilization. Annual patterns of NH₃ fluxes correlate strongly with air temperature (Todd et al., 2011b). Comprehensive chemical surveys confirmed that both the highest overall odor emission rates and the highest total odorous VOC emission rates from livestock barns consistently occurred during the warmer spring and summer seasons (Akdeniz et al., 2012; Cai et al., 2015).

The physical and chemical mechanisms governing emissions from manure itself proved to be temperature-dependent. For example, researchers found that ammonium sorption by feedyard manure was 112% higher at cold temperatures (4 °C) compared to warm conditions (22 °C), meaning that winter weather naturally and temporarily reduced NH₃ volatilization by enhancing manure’s capacity to retain ammonium (NH₄⁺) ions (Waldrip et al., 2012).

4.1.2.3 Identifying hazardous spikes during specific operational events

Temporal emission tracking also proved necessary for identifying extremely hazardous, short-term spikes in pollutants caused by routine farm-management activities. These operational events could create dangerous conditions for both workers and animals if not properly anticipated and managed. For example, slurry agitation and removal operations in deep-pit barns were found to increase H₂S emissions by nearly 62 times relative to baseline concentrations, revealing the critical need for maximum temporary ventilation rates to prevent lethal gas accumulation in confined spaces (Hoff et al., 2006).

Similarly, routine animal bedding events in cattle barns were proven to cause TSP concentrations to spike dramatically, with levels increasing from approximately 58 μg/m³ to over 700 μg/m³ during active bedding distribution (Spiehs et al., 2014c). However, temporal tracking demonstrated that these elevated concentrations returned to baseline levels within 18 h, providing critical information for establishing safe re-entry times and work scheduling protocols. These findings underscored that emission measurements taken only during dormant periods would fail to capture the full range of worker and animal exposure risks, making event-based monitoring essential for occupational health and safety assessments.

4.1.3 Applied quantification: validating models, methods, and abatement solutions

In the most recent stages of emission research, the focus shifted toward evaluating complex computer simulations that can predict dynamic emissions based on localized farm climate and management practices. Computer system-level simulations allowed researchers to overcome the time and field constraints imposed by expensive replicated studies, and robust quantification methods developed in earlier phases enabled this transition; shifting towards active abatement evaluation in a virtual simulated environment, quantifying precisely how determined physical interventions and facility designs would suppress or alter gas and PM fluxes.

4.1.3.1 Validating process-based biogeochemical and atmospheric dispersion models

Recognizing the persistent flaws in static emission factor approaches used in regulatory assumptions, researchers validated complex biogeochemical models. For instance, studies in Texas Panhandle proved that tools like Manure-DNDC (QFLi et al., 2012) and the Integrated Farm Systems Model (IFSM; Rotz et al., 2022) could predict daily NH3 fluxes with 71 to 81% agreement with measured values by dynamically accounting for variables such as air temperature and dietary crude protein content (Waldrip et al., 2014; Waldrip et al., 2013). In contrast, these studies demonstrated that the constant EFs used by the U.S. EPA drastically underestimated actual local feedyard emissions by as much as 79%, highlighting the inadequacy of simplified regulatory approaches.

Building upon those findings, researchers iteratively refined whole-farm models to work across different climatic regions and facility types. Revised versions of IFSM were validated to predict both NH₃ and N₂O fluxes simultaneously for arid Texas beef feedyards and Idaho open-lot dairies, though accuracy varied between regions, ranging from 59 to 81% agreement at Texas sites compared to 92% at the Idaho location (Bonifacio et al., 2015b). To further improve model performance, advanced empirical frameworks that integrated the complex chemical properties of manure itself were developed. For example, incorporating ultraviolet–visible (UV–vis) spectral characteristics of organic matter stability into predictive models yielded 52 to 61% agreement with field observations (Waldrip et al., 2017). The UV–vis absorbance ratio was used to characterize key biochemical properties of manure organic matter, including lignin and aromatic and phenolic compounds containing carbon–carbon double bonds. These properties influence the rate of gaseous emissions, and their incorporation into the model was shown to improve predictive accuracy under high emission scenarios.

While process-based models focused on predicting emission generation at the source, atmospheric dispersion models addressed the complementary challenge of quantifying how emitted pollutants transport and disperse downwind from livestock facilities. Atmospheric dispersion modeling applications grew substantially since the 1990s, and inverse dispersion modeling gained traction in the mid-2000s as a practical tool for quantifying emissions from open-lot livestock facilities. This approach allowed researchers to calculate emission fluxes by measuring downwind pollutant concentrations along with meteorological parameters such as wind speed and atmospheric stability, rather than attempting direct measurement of the area source of emissions, which could be as large as 60–100 ha and contain a wide variety of pen sizes, stocking rates, animal growth stages, and feeding regimes. As researchers increasingly adopted reverse dispersion modeling to calculate emission fluxes, it became necessary to compare the leading software frameworks to ensure consistency and establish methodological standards. Comparisons between the U.S. EPA’s preferred Gaussian-based AERMOD (Cimorelli et al., 2005) and backward Lagrangian stochastic (bLS) models (Flesch et al., 1995) like WindTrax (Thunder Beach Scientific, 2020) revealed that AERMOD consistently derived PM₁₀ emission fluxes 32 to 69% higher than WindTrax for the same field datasets, prompting researchers to propose conversion factors between the two systems (Bonifacio et al., 2013a).

When applied to real-world regulatory evaluations, reverse modeling using AERMOD demonstrated that actual PM₁₀ EFs at large commercial feedlots, measured at 27 to 30 kg per 1,000 head-day, were significantly smaller than the published, generalized U.S. EPA EFs for cattle feedlots (Bonifacio et al., 2012). These findings illustrated how site-specific inverse modeling could provide more practical emission estimates than applying broad regulatory defaults, offering promise for improving the accuracy of emission inventories while also revealing the continued need for model refinement.

4.1.3.2 Evaluating abatement solutions and facility design

Abatement and mitigation research expanded significantly in the early 2000s (Figure 5), driven by the need for practical interventions to reduce emissions. Quantification studies played crucial roles in evaluating and determining which promising solutions could be effectively adopted by producers. For instance, actively injecting swine manure into the subsurface rather than relying on traditional surface spreading reduced the flux of highly odorous aromatic compounds by 80 to 95%, with fluxes dropping below detection limits within 4 to 8 h (Parker et al., 2013a; Parker et al., 2013b). Similarly, incorporating beef cattle manure into cropland using active disk tillage significantly reduced emissions of principal VOC odor compounds compared to leaving manure on the soil surface (Woodbury et al., 2016).

However, quantification studies also uncovered unintended C- or N-based chemical pollutant trade-offs that complicated mitigation strategies. Quantification studies revealed complex pollutant trade-offs across emission categories (detailed in Section 6.2.1), demonstrating that interventions effective for one pollutant class could affect others, at times in perverse ways. Beyond physical application methods, researchers quantified the direct impact of application volume, establishing that CH₄ and CO₂ emissions from cropland scaled systematically with the amount of manure applied, with the greatest emissions resulting from an 80 t/ha application rate (Phan et al., 2012) as compared to application rates of 0 and 40 t/ha.

Beyond testing active interventions applied to manure, quantification studies also evaluated how fundamental facility architecture and existing management systems influence baseline emissions, providing insights for designing lower-emission animal facilities from the outset. In poultry production, researchers established that transitioning from traditional high-rise laying hen houses to manure-belt systems drastically reduced NH₃ emissions. High-rise systems, where manure accumulated in ground-level floor beneath caged hens housed in the upper floor for extended periods, produced substantial emission rates of 0.81 to 0.90 g/d per hen, whereas manure-belt houses using frequent mechanical manure removal reduced those emissions to 0.054 to 0.094 g/d per hen (Liang et al., 2005). This reduction, i.e., 90–93%, demonstrated that manure management frequency and minimizing manure exposure time were critical design considerations.

Conversely, some structural comparisons revealed negligible differences, highlighting that not all design variables influenced emissions equally. When evaluating swine finishing barns, comprehensive monitoring proved that there was no statistically significant difference in airborne PM or bioaerosol concentrations between naturally ventilated and mechanically ventilated housing systems, indicating that ventilation style alone was not a sufficient abatement tool for PM control (Predicala et al., 2001). These comparative facility assessments underscored the importance of empirical quantification in distinguishing which design elements genuinely reduced emissions versus those that had minimal practical impact. Concerns were also raised about the inherently inferior design of a typical swine barn with mechanical ventilation, the so called ‘tunnel barn’, where air enters on one short side of the building and exits from a bank of fans on the other short side end. This design requires more energy to properly ventilate compared with the ‘chimney’ type ventilation where the natural buoyancy of warming up air is boosted by exhaust fans mounted in the ceiling and ejecting the barn air vertically and well above ground level. This over-the-roofline placement of barn exhaust air helps with dispersion at the source and thus limits the range of downwind impact. It also enables a more uniform air quality inside the barn compared with the tunnel system where the exhaust zone end of barn is of most concern. The chimney barn design is more common in the EU compared with North America.

5 Abatement solutions

Abatement solutions represent the culmination of livestock emission research, providing practical strategies to address the environmental challenges faced by industry. Research attention across species has been uneven, with cattle and dairy receiving the highest number of studies, followed by swine and poultry (Tables 1–3). Because abatement efficacy frequently differs between controlled and operational settings, each technique in Tables 1–3 is labeled by study scale (laboratory, field, or both). Interestingly, abatement methodologies have shown consistency across different species, with similar approaches being tested in parallel across livestock types. For example, alternative feed strategies using distiller grain byproducts have been evaluated in both cattle and swine to determine whether these cost-effective feed options influence odor and NH₃ emissions from animal feces (Spiehs et al., 2012a, 2012b). The diverse array of abatement solutions can be organized into three primary categories based on the intervention point: diet manipulation, facility modification, and manure management. Facility modification and manure management represented the major abatement categories. Although diet manipulation showed a higher proportion of studies conducted at the field scale, this likely reflects the need to evaluate dietary interventions in live animals, whereas bedding and manure treatments can often be tested under controlled laboratory or pilot scale conditions (Figure 6).

Table 1. Abatement solutions for cattle and dairy emission control and their efficacy.

Bold text indicates a significant treatment effect, whereas regular text indicates a non-significant effect.

Table 2. Abatement solutions for swine emission control and their efficacy.

Bold text indicates a significant treatment effect, whereas regular text indicates a non-significant effect.

Table 3. Abatement solutions for poultry emission control and their efficacy.

Bold text indicates a significant treatment effect, whereas regular text indicates a non-significant effect.

5.1 Approaches in abatement solutions

5.1.1 Diet manipulation

Diet manipulation strategies in livestock production serve dual purposes: evaluating whether alternative, lower-cost feeds introduce additional emissions and deliberately modifying feed composition to reduce emission levels, while still maintaining production targets. In the area of alternative feeds, distiller’s dried grains with solubles (DDGS), a byproduct of ethanol production, have been tested in both cattle and swine operations (Spiehs et al., 2012b; Spiehs et al., 2012a). Interestingly, the effects of DDGS inclusion varied by species, with cattle manure showing increased NH₃, P, and S concentrations, while swine manure composition remained largely unaffected. Beyond alternative feed evaluation, intentional dietary modifications have been explored to minimize emissions at the source. Archibeque et al. (2006a) hypothesized that substituting steam-rolled corn for dry-rolled corn could enhance starch digestibility in cattle, thereby reducing odorous compounds in the resulting manure. The results suggested that feeding steam-rolled corn reduced fecal starch and odorous compounds with no significant impact on animal productivity. In poultry production, the selection of specific nutrient sources has demonstrated significant impacts on emission profiles. Chavez et al. (2004b) found that the methionine source included in broiler feed influenced the types of VOCs emitted from manure. Across these studies, diet manipulation reduces emissions at the source rather than after excretion, which avoids the pollutant-swapping seen with surface treatments (Section 5.4.3). Its main limitation is that effects are species-specific and modest in magnitude.

5.1.2 Facility modification

Animal facility modifications represent another critical intervention point for emission abatement, with strategies varying considerably across livestock species based on their respective housing systems. In cattle operations, research has concentrated extensively on bedding materials, evaluating different substrate types for their capacity to reduce GHGs and odor emissions. Among the materials tested, pine chips have emerged as particularly effective, demonstrating the ability to reduce NH₃ emissions by up to 14% when mixed at concentrations exceeding 20% with corn stover (Spiehs et al., 2016). However, this approach presents certain tradeoffs, as high proportion of pine chips in the bedding has been associated with increased total reduced sulfide concentrations in ambient air (Spiehs et al., 2016), and practical implementation may be constrained by cost and availability. The divergence in housing systems across species has led to fundamentally different abatement approaches, with cattle operations focusing primarily on pen and bedding management due to their outdoor or open facility designs, while swine and poultry production has concentrated on air filtration and scrubbing technologies suited to their enclosed, mechanically ventilated buildings. Biofilters are among the most consistently effective housing technologies for swine, reducing odor, NH₃, and H₂S simultaneously across multiple independent studies (Table 2). Unlike topical additives, their effect is validated at field scale rather than only in the lab. It is important to note that evaluation periods for these technologies in swine and poultry operations have varied substantially across studies, ranging from as brief as 4 d (Jerez et al., 2013) to as long as 13 weeks (Chen et al., 2008).

5.1.3 Manure management

Manure management has attracted the highest level of research interest among the three abatement approaches. A prominent research focus within this category involves the development and testing of manure additives designed to reduce NH₃, odor, and VOCs in cattle, swine, and poultry operations. This field has garnered particularly substantial attention in swine production systems. Additives act through two main mechanisms. N-(n-butyl) thiophosphoric triamide (NBPT) works by inhibiting urease to reduce NH₃, while treatments such as soybean and calcium peroxidase target odorous VOCs. Reported efficacies vary widely (Table 2), but this variation reflects the scale of experiment more than the technology itself. Reductions above 90% in sealed laboratory vessels fall to roughly 20% on commercial pens (Maurer et al., 2017a).

Across cattle, swine, and poultry production, manure additive research has concentrated primarily on NH₃ emissions and odor control (see Tables 1–3). The overall pattern suggests that while research-grade manure additives show consistent effectiveness for gases such as H₂S, their performance in reducing NH₃ remains more variable. Novel surficial manure treatments with biochar treat both NH₃ and H₂S, and a suite of odorous VOCs, but they still need to be tested at farm scales (Chen et al., 2024). Potential added benefit to biochar treatment of manure is the ‘soaking’ of nutrients to biochar or pretreating feedstock to biochar that could be manipulated as a slow-release fertilizer, in addition to early lab-scale evidence of NH₃ and H₂S emissions mitigation from land-applied swine manure (O’Brien et al., 2026). It should be noted that commercialized manure treatment products for swine have not been showing any sustained and cross-referenced performance (Chen et al., 2020b; Chen et al., 2020a). Treatment approaches have shown some species-specific preferences, with NBPT and soybean peroxidase representing common techniques in cattle and swine operations, whereas poultry research has increasingly shifted toward zeolite-based interventions. Poultry and swine barn operations may benefit from UV-light based technology that is steadily showing readiness for farm-scale trials (Lee et al., 2022). Added potential benefit is the potential bactericidal effect of UV for treating perennial viruses such as PRRS (Li et al., 2022).

5.2 Emission priorities across livestock species

Research attention has concentrated unevenly across pollutant types and livestock species. The common emissions of interest include NH₃, odorous VOC, GHG (CO₂, N₂O, CH₄), and other odorous gases such as H₂S.

In cattle and dairy operations, research has prioritized excreta-related odor emissions (Spiehs et al., 2017; Spiehs et al., 2013) and gaseous NH₃ from feces and urine (Cole et al., 2003; Parker et al., 2016b; Shi et al., 2001; Todd et al., 2006; Waldrip et al., 2015). The open-air nature of feedlot systems has introduced an additional concern largely absent from other livestock sectors: PM emissions from feedlot surfaces (Bonifacio et al., 2012; Bonifacio et al., 2011; Bush et al., 2014; Razote et al., 2006). This PM dispersion distinguishes cattle emission research from that of other livestock species.

Swine production research has centered on emissions from manure storage infrastructure, including lagoons and handling systems (Hudson et al., 2007b; Hudson et al., 2001; Parker et al., 2012a). Additionally, the enclosed, mechanically ventilated nature of modern swine facilities has generated substantial research interest in biofilter technologies to improve indoor air quality (Chen et al., 2009; Chen et al., 2008). This architectural characteristic enables air filtration strategies impractical for open-air systems, fundamentally shaping the trajectory of swine emission abatement research to the end-of-the-pipe type.

Poultry operations have attracted research attention primarily for odor (Cai et al., 2007; Chavez et al., 2004b; Kalus et al., 2017), H₂S (Jerez et al., 2013), and NH₃ emissions (Laor et al., 2010; Singh et al., 2009) from manure. Similar to swine production, the enclosed housing systems common in poultry operations have facilitated research on contained emission control technologies, though the litter-based manure management presents distinct challenges compared to liquid manure systems.

5.3 Laboratory versus field testing: the implementation gap

Laboratory testing typically serves as the initial step in evaluating newly proposed abatement techniques, offering researchers the ability to establish proof of concept under controlled conditions with greater speed and precision than field-scale studies. The controlled environment of laboratory settings enables more accurate measurement of abatement effectiveness by minimizing confounding variables associated with environmental variability, inconsistent management practices, and the inherent complexity of full-scale production systems. Biosecurity protocols in commercial settings make the initial lab- and pilot-scale testing necessary before committing resources and elevating risks at farm-scale. This methodological and cost advantage has made laboratory testing the preferred approach for preliminary evaluation across diverse abatement strategies, including feedyard surface amendments to minimize emissions (Parker et al., 2005; Shi et al., 2001), biofilter effectiveness (Almuhanna et al., 2009) and bedding material performance in controlling odor and VOCs (Spiehs et al., 2016; Spiehs et al., 2013). However, evaluating whether laboratory-observed reductions translate to field conditions requires accounting for spatial and environmental variability that controlled settings cannot replicate, as examined comprehensively in Section 6.2.2. This laboratory-to-field implementation challenge is evident in research identifying critical feedlot surface moisture content of 20% for reducing PM emissions (Guo et al., 2011b), where translating this precise threshold into commercial practice requires further investigation into cost-effective monitoring approaches and adaptive management strategies that can operationalize laboratory findings at field scale (Lee et al., 2026).

5.4 Performance and feasibility of abatement technologies

5.4.1 Economic feasibility of abatement technologies

Economic feasibility fundamentally determines whether abatement technologies will be adopted, as environmentally effective solutions that impose prohibitive costs remain impractical for commercial operations. Some studies have provided detailed economic analyses, such as per-pen installation costs for soil cement feedyard surfaces (Parker et al., 2004), offering producers concrete financial metrics for decision-making. However, cost reporting remains inconsistent across the abatement-focused literature, particularly for consumable-based solutions where total expenses depend on application frequency and site-specific management conditions. A comprehensive review of abatement measures in livestock production revealed the limited extent of economic reporting: only 16 of 94 studies on swine manure management and 17 of 106 studies on livestock housing provided cost estimates in any form (Maurer et al., 2016). This means that approximately 83% of manure management studies and 84% of housing studies omitted economic data entirely, leaving substantial gaps in information necessary for practical implementation decisions. Available cost data also illustrate substantial variability in reporting approaches: NBPT applications for NH₃ reduction are reported at $0.12 per application per head (Shi et al., 2001), while soybean peroxidase treatments are expressed as approximately 1.5% of marketed hog value (Maurer et al., 2017a). These incompatible units complicate direct cost comparisons across abatement strategies. Furthermore, even when cost estimates are provided, they rely on numerous underlying assumptions that introduce additional variability, including the economies of scale. For instance, Miller et al. (2005) estimated costs for H₂S reduction using pelleted feed based on assumptions about average pig weight gain, weight at market, and the price differential between pelleted and ground feed, all of which fluctuate with market conditions. Consequently, economic analyses reported in research findings should be interpreted cautiously when comparing cost effectiveness across different studies or production contexts. Understanding initial costs represents only the first step in economic evaluation, as the total expense of abatement strategies depends on how long these solutions remain effective.

5.4.2 Longevity of abatement effects

Beyond initial implementation costs, the longevity of abatement effectiveness represents an important determinant of economic feasibility, as the duration of effective performance directly influences recurring costs and the ultimate cost per animal produced. Although manure additives are considered relatively inexpensive and can be integrated into existing management systems without extensive operational modifications (Maurer et al., 2016), the frequency of required reapplication substantially influences total implementation costs. Frequent reapplication requirements, such as NBPT treatments needed every 4 d (Parker et al., 2012c), accumulate the per-application costs noted in Section 5.4.1. These recurring expenses accumulate rapidly over production cycles, transforming seemingly modest per-application costs into substantial annual abatement expenditures. Understanding temporal performance patterns enables producers to calculate annualized costs and compare different abatement strategies on an equivalent basis. However, longevity data remains limited across many abatement technologies, with evaluation periods varying substantially among studies. Some research has documented specific performance durations, such as biofilters for swine facility air filtration that maintained 87% effectiveness over 37 d (Ro et al., 2008). However, other studies have documented rapid declines in abatement effectiveness over time. Parker et al. (2016a) applied soybean peroxidase on a lab scale to reduce odorous VOCs from swine manure slurry and found that odor returned by day 14, with emissions exceeding control levels, leading the researchers to recommend reapplication intervals of 4 to 7 d. Additionally, many studies have acknowledged the need for extended evaluation periods, explicitly noting that long-term efficiency data are essential for assessing practical viability (Bonifacio et al., 2011; Razote et al., 2006). Jerez et al. (2013) evaluated this air treatment technology for reducing PM, NH₃, and H₂S, measuring effectiveness over only two consecutive days in fall and winter seasons. Despite providing a detailed 10-y economic analysis that included maintenance costs estimated at 2% of fixed costs, the projections necessarily assumed that equipment effectiveness would remain stable throughout the decade-long period. This gap in long-term performance data represents a significant barrier to determining economic feasibility, as reliable information about sustained effectiveness over multiple production cycles is essential for evaluating whether abatement investments are economically justifiable.

5.4.3 Pollutant trade-offs and unintended consequences

An important dimension of abatement effectiveness is whether reducing target emissions inadvertently increases other pollutant emissions. Table 4 synthesizes the documented trade-offs across abatement technologies. Some researchers have explicitly addressed this concern: Rockafellow et al. (2012) emphasized the necessity of verifying that UV treatment of livestock barn exhaust does not generate ozone (O_₃) and N₂O as treatment byproducts and later addressed by Lee et al. (2022). One approach to managing pollutant trade-offs involves identifying optimal application thresholds where target emission reductions are maximized while minimizing unintended secondary pollutant increases. Spiehs et al. (2016) demonstrated this strategy in evaluating pine chip bedding materials for cattle operations, where NH₃ reductions came at the cost of increased total reduced sulfides. By systematically testing bedding mixtures ranging from 0 to 100% pine chips, the researchers identified an optimal threshold of 30 to 40% pine chip content that achieved substantial NH₃ reductions while maintaining total reduced sulfides at relatively low levels. Comprehensive multi-pollutant monitoring is therefore essential for determining whether abatement technologies achieve genuine environmental improvements or merely redistribute pollution across different emission categories.

Table 4. Documented pollutant trade-offs in livestock abatement, where reducing a target pollutant increased a secondary pollutant.

6 Discussion

6.1 Measurement technology as the foundation for quantification accuracy

As detailed in Section 3.1.1.1, resolving sample storage biases through innovations like SPME fibers and metallized FEP bags was essential for establishing accurate baseline emissions, enabling all subsequent quantification studies. Without solving the sample storage bias problem, early quantification efforts would have systematically underestimated actual emissions. Standardized calibration tools are needed to prevent emission factor miscalculations while increasing emission quantification accuracy. However, developing these tools is unfeasible without having more sensible characterizations of the environment, particularly airflow and wind patterns. Similarly, standardized calibration tools like Fan Assessment Numeration System (FANS) address systematic biases in ventilation rate measurements, ensuring emission factors reflected operational reality rather than idealized equipment specifications. The refinement of calibration innovations like the FANS provided reliable, continuous airflow measurements, ensuring EFs were rooted in reality rather than theoretical fan capacity (Gates et al., 2005). Similarly, researchers found that wind tunnels and flux chambers heavily underestimated actual VOC fluxes in the field unless calibrated with water evaporative flux ratio correction factors or specifically adjusted air velocities (Parker et al., 2010a; Parker et al., 2013a; Parker et al., 2013b). These measurement technology advances were not merely technical improvements but prerequisites for advancing emissions science. The integrity of subsequent quantification studies establishing baseline EFs, and even the evaluation of abatement solution effectiveness, depended entirely on the accurate measurement technologies developed during this early period. Figure 7 illustrates this interdependence concretely, tracing a single pollutant (feedlot PM) from the measurement innovations that made reliable quantification possible, through the identification of surface moisture as the dominant emission driver, to the water-application abatement that this quantified relationship enabled.

6.2 Quantification studies revealing abatement complexity and trade-offs

6.2.1 The moisture trade-off: dust suppression vs. spikes in odor and greenhouse gases

Water applications represent a common abatement strategy used to suppress PM emissions from livestock facilities. Field quantification studies demonstrated its effectiveness, with maintaining 20 to 30% pen surface moisture content reducing both TSP and PM₁₀ emissions by up to 60% (Bonifacio et al., 2015a). However, quantification research examining environmental drivers of emissions revealed that altering this single environmental condition creates significant pollutant trade-offs that would not be apparent without comprehensive monitoring. Studies tracking how moisture shifts emission chemical profiles found that under wet conditions, highly offensive sulfide compounds (Woodbury et al., 2014) contributed 87.3% of the total Odor Activity Value (Woodbury et al., 2015). Furthermore, applying water to dry manure surfaces triggered substantial spikes in GHGs, with N₂O fluxes reaching 99.2 mg/m²/h compared to nearly 0 mg/m²/h when no water is applied (Aguilar et al., 2014b). This example illustrates how quantification studies evaluating emissions under varying environmental conditions exposed pollutant trade-offs that single-condition abatement testing failed to capture. These findings demonstrate that evaluating abatement solutions requires systematic, multi-pollutant monitoring approaches rather than focusing on single target emissions. Future abatement development should comprehensively test new protocols that assess whether interventions inadvertently increase secondary pollutants. When complete elimination of all emission categories proves impractical, decision frameworks are needed to prioritize mitigation efforts based on relative environmental and health impacts, or to quantitatively assess trade-offs between different abatement strategies and the unintended emissions they may generate.

6.2.2 The laboratory-to-field efficacy gap

The transition from laboratory proof-of-concept to field-scale implementation reveals a persistent efficacy gap driven by spatial and environmental variability that laboratory conditions cannot replicate. Abatement technologies often demonstrate exceptional promises under strictly controlled, single-condition laboratory tests. For example, applying a renewable biocatalyst combining soybean peroxidase and calcium peroxide to swine manure yielded NH₃ reductions of up to 67.6% alongside substantial odorant reductions in pilot-scale testing (Maurer et al., 2017b). However, when the same additive was evaluated under complex, real-world farm conditions, its NH₃ mitigation effectiveness declined dramatically to just 21.7% (Maurer et al., 2017a). This implementation gap stems largely from extreme spatial and environmental variability inherent to commercial livestock operations that cannot be captured in controlled laboratory settings. Parker et al. (2016b) illustrated this challenge when evaluating urease inhibitors. While the additives performed well in chamber simulations, field application to a commercial open lot feedyard showed no statistical difference in emissions across treatment rates. This occurred because natural spatial variability of NH₃ flux within a single pen ranged from 512 to ~15,000 μg/m²/min, completely masking any treatment effect. The researchers calculated that more than 500 individual flux chambers per pen would be necessary to detect a 10% difference in treatment means. These findings highlight persistent measurement technology and methodological gaps that constrain verification of abatement effectiveness in operational settings. The challenges of quantifying treatment effects amid high spatial variability represent a critical barrier to technology adoption, as producers require convincing evidence of field-scale performance before investing in abatement solutions. Addressing this gap necessitates development of measurement methodologies capable of detecting genuine treatment effects despite the inherent variability of commercial livestock facilities.

6.3 Abatement solutions evolution: from simplistic controls to managing pollution swapping

Early agricultural air quality research frequently focused on neutralizing single nuisance compounds in strictly controlled environments. This reductionist approach created an illusion that simple, topical chemical amendments could solve complex emission problems. For instance, initial laboratory trials applying zeolite to simulated poultry manure vessels showed promising linear reductions in specific odorants like acetic acid and skatole by up to 67% (Cai et al., 2007). However, as researchers advanced to testing cumulative emissions under more realistic conditions, they discovered that manure biogeochemistry is highly complex. Applying higher zeolite concentrations of 5.0% became less effective at reducing N losses than lower doses, demonstrating that simplistic “more is better” chemical interventions do not translate linearly to emission control (Waldrip et al., 2015).

As abatement technologies transitioned from laboratory to field settings, researchers learned that commercial feedlots and barns are governed by dynamic environmental drivers that easily overwhelm basic surface treatments. Long-term, continuous monitoring using advanced atmospheric dispersion models revealed that daily and seasonal environmental shifts dictate emission rates more powerfully than surface treatments. Temperature emerged as a primary driver, with researchers modeling NH₃ fluxes discovering that emissions follow enzymatically catalyzed Arrhenius kinetics. Summer NH₃ fluxes can naturally exceed 100 kg/ha/d, double or triple winter rates, meaning that surface treatments evaluated under mild conditions will be overpowered by the thermodynamic realities of commercial feedlots during peak summer temperatures (Todd et al., 2013; Todd et al., 2011b).

The recognition of pollutant trade-offs and environmental variability fundamentally shifted abatement research philosophy from seeking “silver bullet” surface chemicals toward holistic, multi-pollutant evaluation frameworks. To prevent unintended consequences, researchers developed comprehensive assessment tools like the Air Management Practices Assessment Tool (AMPAT), which enables stakeholders to objectively evaluate mitigation practices against their simultaneous impacts on odors, VOC, PM, and GHGs before implementation (Maurer et al., 2016). Meta-analyses of modern research concluded that the most effective approach to balance local air quality with global climate impacts involves source reduction, stopping pollution before it is excreted. Improving overall feed efficiency and precisely managing dietary nitrogen and energy intake has been identified as offering the greatest opportunity to simultaneously mitigate both NH₃ and enteric CH₄ per unit of livestock product, avoiding the risks of surface-treatment pollution swapping entirely (Liu et al., 2017b). When exhaust treatment remains necessary, the field has increasingly favored biological, system-based solutions. Properly managed wood-chip biofilters exemplify this systems-based approach, capable of simultaneously addressing multiple chemical pathways across odor, H₂S, and NH₃ (Section 5.1.2) without relying on chemical additions that risk generating secondary pollutants. This evolution from single-compound, reductionist interventions toward integrated, multi-pollutant management strategies reflects the maturation of livestock emission abatement science driven by insights gained through comprehensive quantification and measurement technology development.

7 Conclusion

Livestock emission research in the southern and central Great Plains from the early 2000s to mid-2010s encompasses a diverse array of production systems, each with distinct emission profiles and research priorities. Cattle and dairy operations have concentrated primarily on fugitive PM and NH₃ emissions due to their open-lot configurations, while swine and poultry research has focused more heavily on odorous VOC and enclosed air quality management. This body of work demonstrates that the three main research pillars are fundamentally interdependent and critically impact one another.

Measurement technology serves as the foundation of reliable emission research. Without resolving measurement challenges, subsequent quantification and abatement studies would have been built on flawed data. Quantification studies proved essential not only for establishing baseline emission inventories required for regulatory compliance but also for revealing temporal dynamics, environmental drivers, and pollutant trade-offs that shaped abatement strategy development. Abatement solutions represent the ultimate objective of livestock emission research, providing practical tools to address environmental concerns while maintaining economic viability for producers. The evolution of abatement research documented in this historical review illustrates a fundamental paradigm shift from simple, single-pollutant interventions toward holistic, system-based approaches focused on a multi-pollutant focus.

Despite substantial progress, critical knowledge gaps persist that constrain the translation of research findings into practical implementation. The laboratory-to-field efficacy gap remains a persistent challenge, with abatement effectiveness frequently declining when transitioning from controlled experiments to commercial operations. Current measurement methodologies often lack the efficiency necessary to detect genuine abatement effects amid this variability. Additionally, economic feasibility assessment remains limited.

Future livestock emission research must prioritize development of measurement technologies and methodologies capable of efficiently evaluating abatement performance under operational conditions. This includes advancing continuous monitoring systems that can capture temporal emission dynamics while remaining economically practical for extended deployment, as well as statistical frameworks that can distinguish treatment effects from natural variability. Equally important is the establishment of systematic decision-support tools that enable stakeholders to navigate pollutant trade-offs when simultaneous elimination of all emission categories proves impractical. Such frameworks must integrate both environmental impact metrics and economic feasibility data to identify optimization strategies that provide the greatest net benefit. Finally, advances in computational power and data acquisition technology present new opportunities for livestock emission research. Looking ahead, we suggest that emerging tools such as remote sensing and artificial intelligence should be systematically integrated to enhance measurement efficiency, expand spatial coverage, and accelerate identification of effective mitigation strategies.

In memoriam

This article is dedicated to the memory of Prof. Calvin B. Parnell, Jr., Professor Emeritus of Texas A&M University.

Generative AI statement

The author(s) declared that Generative AI was used in the creation of this manuscript. During the preparation of this work the author used Google Scholar Labs, NotebookLLM, and Claude in order to search literature, edit, and grammar-check the original manuscript. After using those tools/services, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Daniela Lovarelli, University of Milan, Italy

Reviewed by: Brendan Higgins, Auburn University, United States

A. V. Anju, National Institute for Interdisciplinary Science and Technology (CSIR), India